반응형 Training1 neural network hyper-parameter 조절하기 neural network 만들면서 chatGPT, Googling 한 정보들을 모아보려고 한다. neural network 구조를 바꿨는데 training loss가 줄지 않았다. hyperparameter를 어떻게 바꾸면 좋을까. ....................아래는 chatGPT 답변 ............................ If your training loss is not decreasing, there could be several reasons for this. Here are some common issues and potential solutions: 1. **Learning Rate:** The learning rate is a critical hyperparameter.. 2024. 2. 7. 이전 1 다음 반응형
